ZnS Quantum Dots Core

Pharmacy, XaVier UniVersity of Louisiana, New Orleans, Louisiana 70125. Received November 11, 2003; Revised Manuscript Received January 8, 2004...
2 downloads 0 Views 331KB Size
NANO LETTERS

Superparamagnetic Fe2O3 Beads−CdSe/ ZnS Quantum Dots Core−Shell Nanocomposite Particles for Cell Separation

2004 Vol. 4, No. 3 409-413

Desheng Wang,† Jibao He,‡ Nista Rosenzweig,§ and Zeev Rosenzweig*,† Department of Chemistry and the AdVanced Material Research Institute (AMRI), UniVersity of New Orleans, New Orleans, Louisiana 70148, Electron Microscopy Facility, Tulane UniVersity, New Orleans, Louisiana 70118, and Department of Pharmacy, XaVier UniVersity of Louisiana, New Orleans, Louisiana 70125 Received November 11, 2003; Revised Manuscript Received January 8, 2004

ABSTRACT This paper describes the synthesis of new nanocomposite nanoparticles that consist of polymer coated γ-Fe2O3 superparamagnetic cores and CdSe/ZnS quantum dots (QDs) shell. A single layer of QDs was bound to the surface of thiol-modified magnetic beads through the formation of thiol−metal bonds to form luminescent/magnetic nanocomposite particles. Transmission electron microscopy (TEM) and energy disperse spectroscopy (EDS) were used to characterize the size, size distribution, and composition of the luminescent/magnetic nanoparticles. Their average diameter was 30 nm with a size variation of ±15%. The nanoparticles were modified with carboxylic groups to increase their miscibility in aqueous solution. A 3-fold decrease in the luminescence quantum yield of the luminescent/magnetic particles and a slight blue shift in their emission peaks compared to individual luminescent QDs were observed. However, the particles were bright and were easily observed using a conventional fluorescence microscope. Additionally, no apparent broadening of the luminescence peak of the QDs could be seen. The luminescent/magnetic nanoparticles were easily separated from solution by magnetic decantation using a permanent magnet. The new particles could be used in a variety of bioanalytical assays involving luminescence detection and magnetic separation. To demonstrate their utility we immobilized anticycline E antibodies on their surface and used the antibody coated particles to separate MCF-7 breast cancer cells from serum solutions. Anticycline E antibodies bind specifically to cycline, a protein which is specifically expressed on the surface of breast cancer cells. The separated breast cells were easily observed by fluorescence imaging microscopy due to the strong luminescence of the luminescent/magnetic nanocomposite particles.

This paper describes for the first time the formation of luminescent and magnetic nanocomposite particles that consist of superparamagnetic core particles (γ-Fe2O3) and a layer of luminescent CdSe/ZnS quantum dots (QDs) on their surface. These particles could be used in a variety of biomedical and biological applications including magnetic separation and detection of cancer cells, bacteria and viruses. CdSe semiconductor quantum dots (QDs) with diameters ranging between 1.5 and 8 nm exhibit strong, tunable luminescence.1-5 They have been widely investigated for their size-dependent optoelectronic properties,6 and for their potential use in optical devices,7 biological labels,8 and sensors.9 Luminescent QDs are brighter, exhibit higher photostability and show narrower emission peaks compared * Corresponding author. † University of New Orleans. ‡ Tulane University. § Xavier University of Louisiana. 10.1021/nl035010n CCC: $27.50 Published on Web 01/24/2004

© 2004 American Chemical Society

to organic fluorophores.8 CdSe QDs are often coated with a layer of ZnS, which has a larger band gap, to passivate their surface and to increase their emission quantum yield through quantum confinement.10 To enable miscibility of the QDs in aqueous solution, their surface has been modified with carboxyl functional groups.11 Alternatively, QDs were encapsulated in water miscible shells such as liposomes,18 silica miscelles,13 and glyconanospheres.14 QDs were also encapsulated in dendrimers15,16 to increase their stability against photooxidation and photobleaching. Luminescent CdSe/ZnS QDs of green emission and red emission colors were synthesized based on a method developed by Peng et al., with minor modifications.4 12.7 mg cadmium oxide and 160 mg lauric acid were mixed in a 100 mL three-neck flask. The mixture was heated to ∼200 °C in a mantle to fully dissolve the cadmium oxide in the lauric acid solution. Then, 1.94 g trioctylphosphine oxide (TOPO) and 1.94 g hexadecylamine (HDA) were added to

Figure 1. Thiol and carboxy modified γ-Fe2O3 beads are reacted with CdSe/ZnS QDs to form the luminescent/magnetic nanocomposite particles.

the solution. The mixture was heated to a temperature higher than 280 °C and kept under a dry nitrogen atmosphere. Upon reaching the desirable temperature the mantle was removed and 2 mL trioctylphosphine (TOP) solution containing 80 mg selenium powder was rapidly injected into the solution under vigorous stirring. It was previously shown that the diameter of the formed QDs depends on the reaction temperature with smaller particles formed at higher temperature.1 The color of the mixture changed from clear colorless to yellow, orange, or red depending on the exact temperature. To form ZnS coating on the CdSe QDs the mixture was cooled to ∼200 °C. Then, after being kept three minutes at this temperature, a solution containing 250 ul hexamethyldisllathiane ((TMS)2S) and 1 mL diethylzinc (Zn(Et)2) premixed in 2 mL TOP was gradually injected into the solution over a minute. The reaction mixture was kept at 180 °C and stirred for 1 h. The solution was cooled to room temperature and the resulting sample of CdSe/ZnS QDs was washed three times with methanol and chloroform. The QDs showed extremely high quantum yield, reaching 100%. We even can see the shining fluorescence under room light. Polymer coated γ-Fe2O3 superparamagnetic magnetic beads were purchased from Indicia Biotech, France. The nanometric γ-Fe2O3 magnetic particles were coated with dimercapto-succinimid acid (DMSA) to stabilize and functionalize the ferrofluid. The surface of the ferrofluid was covered with free thiol (SH) and carboxyl (COOH) residues (3:20 thiol to COOH ratio) to enable covalent coupling of various ligands to the magnetic particles.17 A TEM image (Figure 2a) of the polymer coated particles shows that their

magnetic core averages 10 ( 15% nm in diameter. The hydrodynamic diameter of the particles that include the polymer layer was found to be around 20 ( 10% nm based on dynamic light scattering measurements. The particles were fully miscible in aqueous solution and no aggregation was observed. The coupling between the luminescent CdSe/ZnS QDs and the magnetic beads was based on thiol chemistry. Thiols (-SH) are probably the most utilized functional groups for stabilizing and modifying CdSe QDs.11,15,16,18,19 Thiol groups form stable bonds with metals on the surface of QDs such as cadmium and zinc.20-24 However, the coupling reaction between the QDs and magnetic beads presented a difficulty since the trioctylphosphine oxide (TOPO) capped CdSe/ZnS QDs were dissolved in chloroform while the polymer-coated magnetic beads were dispersed in water. We found that running the coupling reaction in a 10:5:1 mixture of chloroform/methanol/water yielded the best nanocomposite particles with minimal aggregation. To carry out the coupling reaction we first transferred 1 mL of 1 µM CdSe/ZnS QDs into a 5 mL vial. Then we added 500 µL methanol to the solution. This was followed by the slow injection of 100 µL 0.1 µM magnetic bead-water suspension to the solution under sonication and vigorous stirring. Under these reaction conditions the molar ratio between the QDs and the magnetic particles was 100:1. The excess of QDs was imperative to preventing aggregation of the magnetic beads. Under vigorous stirring the aqueous and organic phases formed an even suspension. The suspension was stirred for 1 h to form the nanocomposite particles. The luminescent/magnetic nanoparticles were then separated from the solution by using a permanent magnet (average cross sectional force density 16.6 T2/m) and washed several times with methanol. The sample was vortexed briefly and sonicated for 15 min to prevent aggregation. A representative TEM image of the nanocomposite particles is shown in Figure 2b. The particles average 20 nm in diameter with a size distribution of about 15% and show minimal or no aggregation. Assuming the magnetic beads are fully covered with QDs, the maximum number of

Figure 2. (a) A TEM image of magnetic beads (γ-Fe2O3) coated with a functionalized polymer; the scale bar is 20 nm. (b) A TEM image of QD-magnetic beads core-shell nanoparticles. The scale bar is 20 nm. (c) A High-resolution TEM image of a single magnetic bead coated with quantum dots. The scale bar is 5 nm. 410

Nano Lett., Vol. 4, No. 3, 2004

Figure 3. EDS (energy disperse spectroscopy) spectrum of a single magnetic bead-QDs core-shell particle.

QDs per magnetic bead could be estimated based on the following equation: N ) 2π (RFe + RQD)2/x3 RQD2 where RFe is the radius of the magnetic beads and RQD is the radius of the smaller QDs. This estimated expression is derived by dividing the surface area covered by the small QDs on the larger iron oxide particle by the area covered by a single QD on the iron oxide particle surface. The calculation assumes close packing of QDs on the magnetic particle surface and takes into consideration the gaps between the QDs. For example, for a magnetic particle diameter of 10 nm, the maximum number of QDs increases from 45 to 133 when the QD diameter decreases from 4 to 2 nm. A high-resolution TEM image of an individual magnetic bead coated with CdSe/ZnS QDs is shown in Figure 2c. Gaps between the QDs can be seen, which indicates imperfect coating. It also implies that the number of QDs per magnetic particle would be lower than the upper theoretical limit. An energy dispersed spectrum (EDS) of the nanocomposite particles is in Figure 3. The Cu peaks result from the sample grid. Spectral peaks that originate from the QD coating of the magnetic beads indicate the presence of Cd, Se, Zn, and S on the surface of the iron oxide particles. The relatively high Zn and S peaks could indicate the presence of multiple layers of ZnS on the surface of the CdSe QDs.25-27 Luminescence images of the nanocomposite particles coated with ∼3 and ∼5 nm CdSe/ZnS QDs are shown in Figures 4a and 4b, respectively. A large signal-to-background ratio of over 100 is observed in these digital images. No micrometric clusters of nanocomposite particles are seen. Luminescence spectra of CdSe/ZnS QDs in chloroform and CdSe/ZnS QD-γ-Fe2O3 nanocomposite particles in aqueous solution are shown in Figure 5. A slight blue shift is observed that could be attributed to a change in surface Nano Lett., Vol. 4, No. 3, 2004

Figure 4. Digital fluorescence microscopy images of magnetic beads coated with (a) 3 nm (green) and (b) 5 nm (red) CdSe/ZnS QDs.

Figure 5. Luminescence spectra. (a, red): CdSe/ZnS QDs in chloroform; (b, green): magnetic beads CdSe/ZnS QD core-shell nanoparticles in water.

states of the QDs due to the immobilization. The emission quantum yield of the nanocomposite particles was found to be around 0.18, which is three times lower than the emission quantum yield of CdSe/ZnS QDs in chloroform (0.61). Luminescence lifetime of the luminescent/magnetic nanoparticles showed an excited-state lifetime of 65 ( 5 ns. The excited-state lifetime of CdSe/ZnS nanoparticles was 27 ( 3 ns. This excited-state lifetime of the CdSe/ZnS QDs was 411

Figure 6. Covalent attachment of anticycline E to the luminescent/magnetic particles using EDAC coupling chemistry.

in agreement with previous studies.28 It should be noted that the lifetime measurements yielded only approximate values since the fluorescence decay times exhibited significant variations from exponential decay curves. Nevertheless, the clear increase in excited state lifetime could be attributed to quenching interactions between the magnetic nanoparticles and the luminescent QDs or between the close packed QDs. The drop in the emission quantum yield could also be attributed to the solvent change and to possible changes in electronic density on the surface of the QDs due to their immobilization to the magnetic particles. While the ZnS capping passivates the surface of the CdSe QDs, the capping is not perfect. This could allow electrons to leak to the surface of the QDs and to the polymer coated magnetic core particle. Leakage of electrons from QDs has been previously observed when the surface of CdSe/ZnS QDs was modified with mercaptoacetic acid (or MPA, MUA).18 This resulted in a decrease in the emission quantum yield and a blue shift in the emission spectrum of the modified QDs as observed in our experiments. It should be noted, however, that the width of the emission peak of the QDs was not affected by the coupling of QDs to the polymer coated magnetic nanoparticles. To apply the luminescent/magnetic particles for magnetic separation of cells followed by luminescence detection, we first modified their surface with carboxylic functional groups. A 2 mL portion of 0.1 µM luminescent/magnetic nanoparticle suspension in methanol was transferred into a 250 mL flask, which was heated to 60 °C under stirring under reflux for 6 h. Then, 200 µL mercaptoacetic acid (Sigma) was injected into the flask. The reaction took 1 h to complete. The carboxyl-modified nanoparticles were washed three times using magnetic decantation and were finally dispersed in 2 mL deionized water and briefly vortexed to prevent aggregation. As expected, the carboxy-modified nanoparticles were highly water-soluble and also maintained their luminescence properties.11,19 Mouse anticycline E antibody molecules were attached to the particles through EDAC (1-ethyl-3-(3dimethylaminopropyl) carbodiimide) coupling as shown in Figure 6. A 10 µL portion of mouse anticycline E antibody (Zymed) solution was added into 1 mL ∼0.1 µM luminescent/ magnetic nanoparticles suspension in a phosphate buffer (PB) solution at pH 7.4. Then 100 mg EDAC coupling reagent was added into the mixture. The mixture was gently shaken for 1 h and the antibody-labeled particles were then separated and washed using magnetic decantation. The anticycline E labeled nanoparticles were resuspended in 1 mL PB solution at pH 7.4. The use of EDAC coupling enabled the covalent attachment of the antibodies to the carboxyl-modified 412

Figure 7. 10× fluorescence (a) and transmission (b) microscopy images of anticycline E labeled luminescent/magnetic nanoparticles bound to MCF-7 breast cancer cells.

particles without substantially damaging the active site of the antibodies. To demonstrate the separation capability of the particles we incubated a sample of 200 µL anticycline E labeled luminescent/magnetic particles with a 1 mL MCF-7 breast cancer cell suspension containing 10,000 cells/mL. Following 15 min incubation at room temperature under gentle shaking the cells were separated from the suspension using a permanent magnet. The cells were washed twice with a phosphate buffer solution using magnetic decantation and observed using digital fluorescence imaging microscopy. Transmission and fluorescence images of the MCF-7 cells labeled with the anticycline E labeled particles are shown in Figure 7. The cells were successfully pulled to the magnet, which meant that they were successfully bound to the anticycline E labeled particles. Control experiments with particles that were not labeled with anticycline E showed negligible nonspecific binding of the particles to MCF-7 cells. While free luminescent/magnetic particles were also Nano Lett., Vol. 4, No. 3, 2004

pulled to the magnet, they could be easily distinguished from cells because of the 3-4 orders of magnitude size difference between the particles and the MCF-7 cells. The cell separation studies described here made use of a single antibody and luminescent QDs of a single emission color. We envision the use of luminescent/magnetic particles of different emission colors and different antibodies where the emission color would code for a specific antibody. This would enable the determination of multiple antigens on the cell surface, which could in turn increase the specificity of cancer diagnosis and staging. Currently we are characterizing the separation capability of the anticycline E labeled particles in mixtures containing MCF-7 and red blood cells. Preliminary measurements indicate that the technique can be used to separate 1 MCF-7 cell in 10 000 blood cells. However, we anticipate that technical improvements in the experimental system would increase the separation efficiency furthermore. Conclusions. Water-soluble nanocomposite particles consisting of a magnetic core (γ-Fe2O3) and luminescent quantum dots shell (CdSe/ZnS QDs) were synthesized in an organic/water two-phase mixture. Thiol chemistry was used to bind the QDs to the surface of the magnetic beads. The luminescent/magnetic nanocomposite particles were characterized using TEM and EDS analysis. The particles averaged 20 nm in diameter with about 15% size variation, showed relatively smooth morphology, and were fully water miscible. They also exhibited high emission quantum yield and were easily separated from solution using a permanent magnet. Anticycline E molecules against the breast cancer specific marker cycline E, were attached to the luminescent/magnetic particles followed their functionalization with carboxylic groups through EDAC coupling. The anti cycline E labeled particles were used successfully to separate and detect breast cancer cells in serum. The separation capabilities of the antibody labeled particles in cell suspensions containing MCF-7 and red blood cells are currently under investigation. Acknowledgment. This work is supported by the National Science Foundation (NSF) grant CHE-0314027 and the Department of Defense grant MDA972-03-C-0100. References (1) Alivisatos, A. P. Science 1996, 217, 933.

Nano Lett., Vol. 4, No. 3, 2004

(2) Murray, C. B.; Norris, D. J.; Bawendi, M. G. J. Am. Chem. Soc. 1993, 115, 8706-8715. (3) Donega´, C. M.; Hickey, S. G.; Wuister, S. F.; Vanmaekelbergh, D.; Meijerink, A. J. Phys. Chem. B 2003, 107, 489-496. (4) Peng, Z. A.; Peng, X. J. Am. Chem. Soc. 2001, 123, 183-184. (5) Cohen, R.; Kronik, L.; Shanzer, A.; Cahen, D.; Liu, A.; Rosenwaks, Y.; Lorenz, J. K.; Ellis, A. B. J. Am. Chem. Soc. 1999, 121, 1054510553. (6) Murray, C.; Kagan, C.; Bawendi, M. Annu. ReV. Mater. Sci. 2000, 30, 546-610. (7) Brus, L. Appl. Phys. A. 1991, 53, 465-474. (8) Warren, C. W.; Nie, S. Science 1998, 281, 2016-2018. (9) Nazzal, A. Y.; Qu, L.; Peng, X.; Xiao, M. Nano Lett. 2003, 3, 819822. (10) Dabbousi, B. O.; Rodriguez-Viejo, J.; Mikelec, F. V.; Heine, J. R.; Mattoussi, H.; Ober, R.; Jensen, K. F.; Bawendi, M. G. J. Phys. Chem. B 1997, 101, 9463-9475. (11) Gerion, D.; Pinard, F.; William, S. C.; Parak, W. J.; Zanchet, D.; Weiss, S.; Alivisatos, A. P. J. Phys. Chem. B 2001, 105, 88618871. (12) Barbara-Guillem, E. US 2002001716 A, 2001. (13) Chen, Y.; Rosenzweig, Z. Nano Lett. 2002, 2, 1299-1302. (14) Chen, Y.; Ji, T.; Rosenzweig, Z. Nano Lett. 2003, 3, 581-584. (15) Wang, Y. A.; Li, J. J.; Chen, H.; Peng, X. J. Am. Chem. Soc. 2002, 124, 2002. (16) Zhang, C.; O’Brien, S.; Balogh, L. J. Phys. Chem. B 2002, 106, 10316-10321. (17) http://www.indicia.fr. (18) Wuister, S. F.; Swart, I.; Driel, F. V.; Hickey, S. G.; Donega´, C. M. Nano Lett. 2003, 3, 503-507. (19) Aldana, J.; Wang, Y. A.; Peng, X. J. Am. Chem. Soc. 2001, 123, 8844-8850. (20) Harruff, B. A.; Bunker, C. E. Langmuir 2003, 19, 893-897. (21) Guo, W.; Li, J. J.; Wang, Y. A.; Peng, X. J. Am. Chem. Soc. 2003, 125, 3901-3909. (22) Wang, Y. A.; Li, J. J.; Chen, H.; Peng, X. J. Am. Chem. Soc. 2002, 124, 2002. (23) Zhang, C.; O’Brien, S.; Balogh, L. J. Phys. Chem. B 2002, 106, 10316-10321. (24) Cohen, R.; Kronik, L.; Shanzer, A.; Cahen, D.; Liu, A.; Rosenwaks, Y.; Lorenz, J. K.; Ellis, A. B. J. Am. Chem. Soc. 1999, 121, 1054510553. (25) Cumberland, S. L.; Berrettini, M. G.; Javier, A.; Strouse, G. F. Chem. Mater. 2003, 15, 1047-1056. (26) Hines, M. A.; Guyot-Sionnest, P. J. Phys. Chem. 1996, 100, 468471. (27) Van Sark, W. G. J. H. M.; Frederix, P. L. T. M.; Van den Heuvel, D. J.; Gerritsen, H. C.; Bol, A. A.; van Lingen, J. N. J.; de Mello Donega, C.; Meijerink, A. J. Phys. Chem. B 2001, 105, 8281-8284. (28) Fisher, B. R.; Hans-Jurgen, E.; Nathan, E. S.; Bawendi, M. G. J. Phys. Chem. B 2003, 108, 143-148.

NL035010N

413